Inferential Statistics Flashcards
1
Q
Inferential Statistics
Criterion of “TRUTH”
A
Validity
2
Q
The percentage of people with the disease who are detected by the test
A
% SENSITIVITY
3
Q
TP ÷ [TP + FN] x 100
A
% SENSITIVITY
4
Q
% SENSITIVITY, higher the sensitivity the better?
A
ye
5
Q
what does % SENSITIVITY measures?
A
TRUE POSITIVE
yung mga tunay na may sakit if ever
6
Q
is the percentage of people with the disease who are not detected by the test, complement of sensitivity
A
% FALSE NEGATIVE
7
Q
FN ÷ [TP + FN] x 100
A
% FALSE NEGATIVE
8
Q
Counterpart of %sensitivity
A
% FALSE NEGATIVE
9
Q
T or F
Higher the sensitivity, the lower the false negative
A
T
inversely proportional sila with %sensitivity
10
Q
is the percentage of people without the disease who are correctly labelled by the test as not diseased.
A
% SPECIFICITY
11
Q
TN ÷ [FP + TN] x 100
A
% SPECIFICITY
12
Q
T or F
Higher the specificity the better – mababa false positive
A
T
13
Q
is the percentage of people without the disease who are incorrectly labelled by the test as having disease, complement of specificity.
A
% FALSE POSITIVE
inversely proportional with %specificity
14
Q
FP ÷ [FP + TN] x 100
A
% FALSE POSITIVE
15
Q
T or F
yes to false positive and false negative
A
F
NO DAPAT
16
Q
is defined as the likelihood that an individual with a positive test has the disease.
A
PREDICTIVE VALUE OF A POSITIVE TEST
17
Q
TP ÷ [TP + FP] x 100
A
PREDICTIVE VALUE OF A POSITIVE TEST
Lahat ng positive result to get who are TRULY POSITIVE
18
Q
is defined as the likelihood that a person with a negative test does not have the disease.
A
PREDICTIVE VALUE OF A NEGATIVE TEST
19
Q
TN ÷ [FN + TN] x 1004
A
PREDICTIVE VALUE OF A NEGATIVE TEST
All of the negative result to get who are FALSE NEGATIVE talaga
20
Q
The ratio of the chance of the test being positive if having the condition compared to the chance of testing positive if not having the condition
A
Positive Likelihood Ratio +LR
21
Q
The ratio of the chance of the test being negative if having the condition compared to the chance of testing negative in not having the condition.
A
Negative Likelihood ratio -LR
22
Q
if u see this card
A
practice the example of maam given for the Indices to Evaluate Accuracy of a Test or Diagnostic Examination
go na
23
Q
Also termed as “reproducibility” or “repeatability”
A
Reliability
Na ulit yung test then same result = reliability – CONSISTENT
Validity = nearest to true value
24
Q
Refers to the stability or consistency of information
A
Reliability
25
The extent to which similar information is supplied when measurements are performed more than once.
Reliability
26
# T or F
A key goal in applied biostatistics is to make inferences about unknown population parameters based on sample statistics.
TRUE
27
what is the difference for parameter and statistics when it comes to mean, SD, and Proportion
Paramerter = Population
Statistic = Sample
## Footnote
this means that kung anong TESTING used for sample, and popluation yun lang din gagamiting sa parameter
28
There are two broad areas of statistical inference,
* Estimation
* Hypothesis Testing
29
The process of determining a likely value for a population parameter (e.g., the true population mean or population proportion) based on a random sample.
Estimation – APPROXIMATION
30
# Estimation - T or F
In practice, we select a sample from the target population and use sample statistics (e.g., the sample mean or sample proportion) as estimates of the unknown parameter
T
31
# Estimation - T or F
The sample should be representative of the population, with participants selected at random from the population.
T
| alam niyo nayan very ez
32
# Estimation - T or F
In generating estimates, it is also important to quantify the precision of estimates from different samples.
T
33
# Estimation
Point Estimate =
Single number
| e.g.: 1, 2, and 69
34
# Estimation
Interval Estimate (Confidence Interval Estimate) =
may decimals (2 values lower and upper limit with confidence intervals)
35
a range of values, derived from sample statistics, that is likely to contain the value of an unknown population parameter.
Confidence Interval
36
# Estimation - confidence interval
Because of their _ _ _ _ _ _ _ _ _ _ _ _ , it is unlikely that two samples from a particular population will yield identical confidence intervals.
Random Nature
37
# Estimation - confidence interval: T OR F
But if you repeated your sample many times, a certain percentage of the resulting confidence intervals would contain the unknown population parameter.
T
| diko parin gets to
38
If you see this card
go over the inferential statistics, check the estimation interval pls
39
There are a number of population parameters of potential interest when one is estimating health outcomes (or "endpoints").
Parameter Estimation
40
# Parameter Estimation
Many of the outcomes we are interested in estimating are either
continuous or dichotomous variables
| , although there are other types.
41
# Parameter Estimation
The parameters to be estimated depend not only on whether the endpoint is continuous or dichotomous, but also on the ?
number of groups being studied.
42
# Parameter Estimation
When 2 groups are being compared what you need to establish between the groups?
* Independent (e.g., men versus women)
* Dependent (i.e., matched or paired, such as a before and after comparison).
43
# Parameters to estimate in health-related studies
One sample - Continuos varible
Mean
44
# Parameters to estimate in health-related studies
One sample - dichotomous variable
Proportion or Rate
| yung mga prevelance,incidence rate ...
45
# Parameters to estimate in health-related studies
2 Independent Samples - Cont. Variable
Difference in MEAN
46
# Parameters to estimate in health-related studies
2 Independents Samples - Dichoto. Variable
Difference in proportion or rates
## Footnote
pag 2 independent samples, lagi difference okay? okay
47
# Parameters to estimate in health-related studies
2 Dependent, Matched Samples - Cont. Variable
Mean Difference
| iba ang difference in means sa mean difference okay? okay
48
# Confidence Intervals
Two types of estimated for each population parameter
* Point estimate
* Confidence interval (CI) estimate.
49
What is the difference between Cont and Dichotomous Variable?
Cont is all about MEAN, while Dicho is proportions or rate
| okay? OKAY
50
# Confidence Intervals
one first computes the point estimate from a sample?
Ye
| para makuha mo Confidence intervals
51
# Confidence Interval - T or F
Sample means and sample proportions are unbiased estimates of the corresponding population parameters.
True
52
If you see this card
go over the PRINCIPLES of confidence interval
| need siya understood, not memorized
53
The confidence interval estimate (CI) is a range of likely values for the population parameter based on
the point estimate, e.g., the sample mean
54
# Confidence Intervals Estimate - T or F
In practice, we select one random sample and generate one confidence interval, which may or may not contain the true mean. The observed interval may over- or underestimate μ.
True
55
# Confidence Intervals Estimate - T or F
The confidence interval does not reflect the variability in the unknown parameter.
T
56
# Confidence Intervals Estimate - T or F
what does confidence interval estimate REFLECTS
amount of random error in the sample and provides a range of values
| likely to include the unknown parameter. sa range of values
57
if u see this card
araling formula sa confidence interval
58
# Confidence Interval
For n >= 30, T or Z table?
Z table
59
# Confidence Interval
For n < 30
Use the t-table with df-n-1
60
# Point Estimate Z SE
where is the Z values got from?
the standard normal distribution for the selected confidence level
| (e.g., for a 95% confidence level, Z=1.96).
61
In practice, we often do not know the value of the population standard deviation (σ).
However, if the sample size is large (n > 30), then the sample standard deviations can be used to estimate the population standard deviation.
Point Estimate (+-) Z SE
62
With smaller samples (n< 30) the Central Limit Theorem does not apply, and another distribution called
T-distribution
63
# Confidence Intervals Estimate for Smaller Samples
Similar to the standard normal distribution but takes a slightly different shape depending on the sample size.
T-distribution
64
# T-Distribution - T or F
In a sense, one could think of the t distribution as a family of distributions for larger samples.
F
| smaller, n <30 - LESSSSSSSSSSSS THAN
65
# T - distribution - T or F
It produces smaller margins of error
F
| It produces LARGER, because small samples are less precise
66
t values are listed by?
degrees of freedom (df)
67
# T-Distribution - T or F
Just as with large samples, the t distribution assumes that the outcome of interest is approximately normally distributed.
T
68
If u see this card
go over the example for Confidence intervals pls pls
| PLEASE
## Footnote
PUHLEASE
69
The sample proportion
p̂ (p hat)
70
confidence interval can be computed by this
p hat formula
| so go over it
71
if u see this card
go over the example of Confidence Intervals Estimate for Population Proportion
72
a contention or assumption made concerning a population characteristics.
STATISTICAL HYPOTHESIS
73
. It is usually concerned with the parameters of the population about which the statement is made.
STATISTICAL HYPOTHESIS
| NOT YET TRUE – ipprove palang
74
The purpose of the research is to provide evidence to support or refute the null hypothesis
Hypothesis Testing
75
Hypothesis testing comprises a set of what?
set of procedures
76
# Hypothesis Testing - T or F
A hypothesis is either rejected or not based on the probability of occurrence of the sample results if the null hypothesis were true.
T
77
how is Statistical Hypothesis validated?
if calculated probability of results exceeds a prespecified value of alpha
78
# Hypothesis
If calculated probability is less than or equal to alpha?
hypothesis is rejected, therefore, result is statistically significant.
| < (less than) = (equal) baka di mo alam eh
79
This is the hypothesis of “no difference”. Statement of equality
Null Hypothesis (Ho)
80
This is the hypothesis of “no relationship”.
Null Hypothesis (Ho)
81
Asserts that population parameter is some value other than one hypothesized.
Alternative Hypothesis (H1 or HA)
82
Usually the research hypothesis, the hypothesis the investigator believes in.
Alternative Hypothesis (H1 or HA)
83
Null hypothesis should always be framed in hopes of being able to reject it so that the alternative hypothesis could be accepted.
oo
84
Include values of statistics leading to rejection of null hypothesis. Usually called alpha or tail of the curve.
Critical Region or Region of Rejection
85
These values are those whose probability of occurrence is less than (<) or equal to the level of sig/nificance, α.
Critical Region or Region of Rejection
86
The probability level that is considered too low to warrant support of the hypothesis being tested.
Level of Significance or ALPHA Level
87
Basis for inferring the operation of non-chance factors (0.05, 0.01, 0.1)
Level of Significance or ALPHA Level
| omegaverse????
88
if u see this card
MASTER the decision table
| yung may Ho true, Ho false
89
# Region of Acceptance
alpha of the curve, greater than or equal to level of significance
1
90
# Region of Acceptance
When Ho is rejected?
statistically significant and the observed difference may not be attributed to sampling variation
91
# Region of Acceptance
If Ho is not rejected
ot statistically significant and may be due to sampling variation
92
what is Ho?
null hypo
93
When HA asserts that population parameter is different from one hypothesized (2-tailed test)
NON-DIRECTIONAL Ha
94
Asserts the direction of the difference ( 1-tailed test)
DIRECTIONAL Ha
95
# Steps in Hypothesis Testing
1st step
Determine whether a 2-tailed or a 1-tailed test be made.
96
# Steps in Hypothesis Testing
2 step, what do you need to assume?
Ho and Ha
| Null and Alternative
97
# Steps in Hypothesis Testing
3rd step, after the hypothesis
Choose alpha, the arbitrary level of significance
| here is the basis of rejection AFTER the computation
98
99
# Steps in Hypothesis Testing
4th and 5th step
5. Determine critical region
6. Determine appropriate test
100
# Steps in Hypothesis Testing
6th and 7th the last
6. Solution
7. Conclusion
101
# Conparison of Parameters or Indicators
Single Population
| what interval/ration testi used
Z or T Test
102
# Conparison of Parameters or Indicators
Single Population
| what ordinal test is used
* Kolmogorov
* SMirnov one sample test
103
# Conparison of Parameters or Indicators
Single Population
| what nominal test is used
* Z test
* Chi Square Test
104
# Conparison of Parameters or Indicators
2 Population: Related Samples
| what interval/ratio test is used
Paired t test
105
# Conparison of Parameters or Indicators
2 Population: Related Samples
| what ordinal est is used
* Wilcoxon
* Matched pairs
* SIgned ranks test
106
# Conparison of Parameters or Indicators
2 Population: Related Samples
| what nominal est is used
McNemar's Test
107
# Conparison of Parameters or Indicators
2 Population: independent Samples
| Interval/Ratio
Independent T-test
108
# Conparison of Parameters or Indicators
2 Population: independent Samples
| Ordinal
Mann whitney U test
109
# Conparison of Parameters or Indicators
2 Population: independent Samples
| Nominal
* fishers exact
* probability test
* Chi square test
110
# Conparison of Parameters or Indicators
3 or More population: Related samples
| interval/ratio
F-test: 2 way analysis of Variance
111
# Conparison of Parameters or Indicators
3 or More population: Related samples
| Ordinal
Friedman's Analysis of Variance
112
# Conparison of Parameters or Indicators
3 or More population: Related samples
| Nominal
Cochran's Q test
113
# Conparison of Parameters or Indicators
3 or More population: Independent
| Nominal
Chi square test
114
# Conparison of Parameters or Indicators
3 or More population: Independent
| Ordinal
Kruskali wallis one way ANOVA
115
# Conparison of Parameters or Indicators
3 or More population: Independent
| Interval/Ratio
F-Test: one way ANOVA
116
# Study of Relationship Between Variables
Interval/Ratio
* Regression
* Correlation
117
# Study of Relationship Between Variables
Ordinal
* Spearman Rank
* Correlation
* Coefficient
118
# Study of Relationship Between Variables
Nominal
* Kappa Test
* Contingenct
* Coefficient Test
119
# What statistical test
if u see this card
go over the relationship for the Independent and dependt and what statistical test will be used
120
# what statistical test is used when
* Independent - Qualitative
* Dependent - Qualitative
Chi square test
121
# what statistical test is used when
* Independent - Qualitative
* Dependent - Quantitative
T,Z, ANOVA
122
# what statistical test is used when
* Independent - Quantitative
* Dependent - Quantitative
Linear Regression
123
# what statistical test is used when
* Independent - Quantitative
* Dependent - Qualitative
Logistic Regression
124
If u see this card
Please go over the CASE, example yan ha
| also read the nte
125
